metadata
annotations_creators:
- crowdsourced
- expert-generated
language:
- en
license:
- apache-2.0
multilinguality:
- monolingual
size_categories:
- 100M<n<1B
task_categories:
- other
pretty_name: P3
dataset_info:
- config_name: adversarial_qa_dbert_answer_the_following_q
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18313753
num_examples: 10000
- name: validation
num_bytes: 1791034
num_examples: 1000
download_size: 6288641
dataset_size: 20104787
- config_name: adversarial_qa_dbert_based_on
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17580553
num_examples: 10000
- name: validation
num_bytes: 1717566
num_examples: 1000
download_size: 6206744
dataset_size: 19298119
- config_name: adversarial_qa_dbert_generate_question
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18552810
num_examples: 10000
- name: validation
num_bytes: 1824231
num_examples: 1000
- name: test
num_bytes: 1954952
num_examples: 1000
download_size: 5882604
dataset_size: 22331993
- config_name: adversarial_qa_dbert_question_context_answer
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 16859685
num_examples: 10000
- name: validation
num_bytes: 1646118
num_examples: 1000
download_size: 6180363
dataset_size: 18505803
- config_name: adversarial_qa_dbert_tell_what_it_is
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17793277
num_examples: 10000
- name: validation
num_bytes: 1739418
num_examples: 1000
download_size: 6276720
dataset_size: 19532695
- config_name: adversarial_qa_dbidaf_answer_the_following_q
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18273217
num_examples: 10000
- name: validation
num_bytes: 1797789
num_examples: 1000
download_size: 6321670
dataset_size: 20071006
- config_name: adversarial_qa_dbidaf_based_on
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17539777
num_examples: 10000
- name: validation
num_bytes: 1724577
num_examples: 1000
download_size: 6247591
dataset_size: 19264354
- config_name: adversarial_qa_dbidaf_generate_question
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18508967
num_examples: 10000
- name: validation
num_bytes: 1830585
num_examples: 1000
- name: test
num_bytes: 1925723
num_examples: 1000
download_size: 5983857
dataset_size: 22265275
- config_name: adversarial_qa_dbidaf_question_context_answer
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 16821505
num_examples: 10000
- name: validation
num_bytes: 1652425
num_examples: 1000
download_size: 6292806
dataset_size: 18473930
- config_name: adversarial_qa_dbidaf_tell_what_it_is
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17755161
num_examples: 10000
- name: validation
num_bytes: 1745717
num_examples: 1000
download_size: 6250903
dataset_size: 19500878
- config_name: adversarial_qa_droberta_answer_the_following_q
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18084393
num_examples: 10000
- name: validation
num_bytes: 1798375
num_examples: 1000
download_size: 6223439
dataset_size: 19882768
- config_name: adversarial_qa_droberta_based_on
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17352073
num_examples: 10000
- name: validation
num_bytes: 1725151
num_examples: 1000
download_size: 6202901
dataset_size: 19077224
- config_name: adversarial_qa_droberta_generate_question
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 18257414
num_examples: 10000
- name: validation
num_bytes: 1828966
num_examples: 1000
- name: test
num_bytes: 1997556
num_examples: 1000
download_size: 5928633
dataset_size: 22083936
- config_name: adversarial_qa_droberta_question_context_answer
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 16638393
num_examples: 10000
- name: validation
num_bytes: 1653815
num_examples: 1000
download_size: 6193786
dataset_size: 18292208
- config_name: adversarial_qa_droberta_tell_what_it_is
features:
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 17571837
num_examples: 10000
- name: validation
num_bytes: 1747043
num_examples: 1000
download_size: 6152157
dataset_size: 19318880
- config_name: ag_news_classify
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 79459523
num_examples: 120000
- name: test
num_bytes: 5007082
num_examples: 7600
download_size: 37504540
dataset_size: 84466605
- config_name: ag_news_classify_question_first
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 79339523
num_examples: 120000
- name: test
num_bytes: 4999482
num_examples: 7600
download_size: 37311664
dataset_size: 84339005
- config_name: ag_news_classify_with_choices
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 91699523
num_examples: 120000
- name: test
num_bytes: 5782282
num_examples: 7600
download_size: 38377186
dataset_size: 97481805
- config_name: ag_news_classify_with_choices_question_first
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 91699523
num_examples: 120000
- name: test
num_bytes: 5782282
num_examples: 7600
download_size: 38318638
dataset_size: 97481805
- config_name: ag_news_recommend
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 94039523
num_examples: 120000
- name: test
num_bytes: 5930482
num_examples: 7600
download_size: 38368116
dataset_size: 99970005
- config_name: ag_news_which_section
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 83899523
num_examples: 120000
- name: test
num_bytes: 5288282
num_examples: 7600
download_size: 37893964
dataset_size: 89187805
- config_name: ag_news_which_section_choices
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 100099523
num_examples: 120000
- name: test
num_bytes: 6314282
num_examples: 7600
download_size: 39167925
dataset_size: 106413805
- config_name: ai2_arc_ARC_Challenge_heres_a_problem
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 870695
num_examples: 1119
- name: validation
num_bytes: 237526
num_examples: 299
- name: test
num_bytes: 929144
num_examples: 1172
download_size: 796298
dataset_size: 2037365
- config_name: ai2_arc_ARC_Challenge_i_am_hesitating
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 1063080
num_examples: 1119
- name: validation
num_bytes: 290313
num_examples: 299
- name: test
num_bytes: 1135794
num_examples: 1172
download_size: 1087298
dataset_size: 2489187
- config_name: ai2_arc_ARC_Challenge_multiple_choice
features:
- name: answer_choices
sequence: string
- name: inputs
sequence: int32
- name: inputs_pretokenized
dtype: string
- name: targets
sequence: int32
- name: targets_pretokenized
dtype: string
splits:
- name: train
num_bytes: 1079865
num_examples: 1119
- name: validation
num_bytes: 294798
num_examples: 299
- name: test
num_bytes: 1153374
num_examples: 1172
download_size: 1096748
dataset_size: 2528037
configs:
- config_name: adversarial_qa_dbert_answer_the_following_q
data_files:
- split: train
path: adversarial_qa_dbert_answer_the_following_q/train-*
- split: validation
path: adversarial_qa_dbert_answer_the_following_q/validation-*
- config_name: adversarial_qa_dbert_based_on
data_files:
- split: train
path: adversarial_qa_dbert_based_on/train-*
- split: validation
path: adversarial_qa_dbert_based_on/validation-*
- config_name: adversarial_qa_dbert_generate_question
data_files:
- split: train
path: adversarial_qa_dbert_generate_question/train-*
- split: validation
path: adversarial_qa_dbert_generate_question/validation-*
- split: test
path: adversarial_qa_dbert_generate_question/test-*
- config_name: adversarial_qa_dbert_question_context_answer
data_files:
- split: train
path: adversarial_qa_dbert_question_context_answer/train-*
- split: validation
path: adversarial_qa_dbert_question_context_answer/validation-*
- config_name: adversarial_qa_dbert_tell_what_it_is
data_files:
- split: train
path: adversarial_qa_dbert_tell_what_it_is/train-*
- split: validation
path: adversarial_qa_dbert_tell_what_it_is/validation-*
- config_name: adversarial_qa_dbidaf_answer_the_following_q
data_files:
- split: train
path: adversarial_qa_dbidaf_answer_the_following_q/train-*
- split: validation
path: adversarial_qa_dbidaf_answer_the_following_q/validation-*
- config_name: adversarial_qa_dbidaf_based_on
data_files:
- split: train
path: adversarial_qa_dbidaf_based_on/train-*
- split: validation
path: adversarial_qa_dbidaf_based_on/validation-*
- config_name: adversarial_qa_dbidaf_generate_question
data_files:
- split: train
path: adversarial_qa_dbidaf_generate_question/train-*
- split: validation
path: adversarial_qa_dbidaf_generate_question/validation-*
- split: test
path: adversarial_qa_dbidaf_generate_question/test-*
- config_name: adversarial_qa_dbidaf_question_context_answer
data_files:
- split: train
path: adversarial_qa_dbidaf_question_context_answer/train-*
- split: validation
path: adversarial_qa_dbidaf_question_context_answer/validation-*
- config_name: adversarial_qa_dbidaf_tell_what_it_is
data_files:
- split: train
path: adversarial_qa_dbidaf_tell_what_it_is/train-*
- split: validation
path: adversarial_qa_dbidaf_tell_what_it_is/validation-*
- config_name: adversarial_qa_droberta_answer_the_following_q
data_files:
- split: train
path: adversarial_qa_droberta_answer_the_following_q/train-*
- split: validation
path: adversarial_qa_droberta_answer_the_following_q/validation-*
- config_name: adversarial_qa_droberta_based_on
data_files:
- split: train
path: adversarial_qa_droberta_based_on/train-*
- split: validation
path: adversarial_qa_droberta_based_on/validation-*
- config_name: adversarial_qa_droberta_generate_question
data_files:
- split: train
path: adversarial_qa_droberta_generate_question/train-*
- split: validation
path: adversarial_qa_droberta_generate_question/validation-*
- split: test
path: adversarial_qa_droberta_generate_question/test-*
- config_name: adversarial_qa_droberta_question_context_answer
data_files:
- split: train
path: adversarial_qa_droberta_question_context_answer/train-*
- split: validation
path: adversarial_qa_droberta_question_context_answer/validation-*
- config_name: adversarial_qa_droberta_tell_what_it_is
data_files:
- split: train
path: adversarial_qa_droberta_tell_what_it_is/train-*
- split: validation
path: adversarial_qa_droberta_tell_what_it_is/validation-*
- config_name: ag_news_classify
data_files:
- split: train
path: ag_news_classify/train-*
- split: test
path: ag_news_classify/test-*
- config_name: ag_news_classify_question_first
data_files:
- split: train
path: ag_news_classify_question_first/train-*
- split: test
path: ag_news_classify_question_first/test-*
- config_name: ag_news_classify_with_choices
data_files:
- split: train
path: ag_news_classify_with_choices/train-*
- split: test
path: ag_news_classify_with_choices/test-*
- config_name: ag_news_classify_with_choices_question_first
data_files:
- split: train
path: ag_news_classify_with_choices_question_first/train-*
- split: test
path: ag_news_classify_with_choices_question_first/test-*
- config_name: ag_news_recommend
data_files:
- split: train
path: ag_news_recommend/train-*
- split: test
path: ag_news_recommend/test-*
- config_name: ag_news_which_section
data_files:
- split: train
path: ag_news_which_section/train-*
- split: test
path: ag_news_which_section/test-*
- config_name: ag_news_which_section_choices
data_files:
- split: train
path: ag_news_which_section_choices/train-*
- split: test
path: ag_news_which_section_choices/test-*
- config_name: ai2_arc_ARC_Challenge_heres_a_problem
data_files:
- split: train
path: ai2_arc_ARC_Challenge_heres_a_problem/train-*
- split: validation
path: ai2_arc_ARC_Challenge_heres_a_problem/validation-*
- split: test
path: ai2_arc_ARC_Challenge_heres_a_problem/test-*
- config_name: ai2_arc_ARC_Challenge_i_am_hesitating
data_files:
- split: train
path: ai2_arc_ARC_Challenge_i_am_hesitating/train-*
- split: validation
path: ai2_arc_ARC_Challenge_i_am_hesitating/validation-*
- split: test
path: ai2_arc_ARC_Challenge_i_am_hesitating/test-*
- config_name: ai2_arc_ARC_Challenge_multiple_choice
data_files:
- split: train
path: ai2_arc_ARC_Challenge_multiple_choice/train-*
- split: validation
path: ai2_arc_ARC_Challenge_multiple_choice/validation-*
- split: test
path: ai2_arc_ARC_Challenge_multiple_choice/test-*
Dataset Card for P3
Table of Contents
Dataset Description
- Homepage: https://bigscience.huggingface.co/promptsource
- Repository: https://github.com/bigscience-workshop/promptsource/
- Paper: Multitask Prompted Training Enables Zero-Shot Task Generalization
- Point of Contact: Victor Sanh
Dataset Summary
P3 (Public Pool of Prompts) is a collection of prompted English datasets covering a diverse set of NLP tasks. A prompt is the combination of an input template and a target template. The templates are functions mapping a data example into natural language for the input and target sequences. For example, in the case of an NLI dataset, the data example would include fields for Premise, Hypothesis, Label. An input template would be If {Premise} is true, is it also true that {Hypothesis}?, whereas a target template can be defined with the label choices Choices[label]. Here Choices is prompt-specific metadata that consists of the options yes, maybe, no corresponding to label being entailment (0), neutral (1) or contradiction (2).
Prompts are collected using Promptsource, an interface to interactively write prompts on datasets, and collect prompt-specific metadata such as evaluation metrics. As of October 13th, there are 2'000 prompts collected for 270+ data(sub)sets. The collection of prompts of P3 is publicly available on Promptsource.
To train T0*, we used a subset of the prompts available in Promptsource (see details here). However, some of the prompts use random.choice, a method that selects uniformly at random an option in a list of valid possibilities. For reproducibility purposes, we release the collection of prompted examples used to train T0*. The data available here are the materialized version of the prompted datasets used in Multitask Prompted Training Enables Zero-Shot Task Generalization which represent only a subset of the datasets for which there is at least one prompt in Promptsource.
Supported Tasks and Leaderboards
The tasks represented in P3 cover a diverse set of NLP tasks including multiple-choice QA, sentiment analysis or natural language inference. We detail the full list of datasets in Source Data.
Languages
The data in P3 are in English (BCP-47 en).
Dataset Structure
Data Instances
An example of "train" looks as follows:
{
'answer_choices': ['safe', 'trolley'],
'inputs': [86, 8, 7142, 666, 6, 405, 8, 3, 834, 1518, 21, 1346, 42, 31682, 58, 37, 3, 929, 9, 3042, 63, 2765, 808, 8, 2045, 6448, 326, 13, 8, 31682, 11, 3, 24052, 135, 16, 8, 1346, 552, 8, 3, 834, 47, 6364, 5], 'inputs_pretokenized': 'In the sentence below, does the _ stand for safe or trolley?\nThe treasury workers took the gold bars off of the trolley and stacked them in the safe until the _ was empty.',
'targets': [31682, 1],
'targets_pretokenized': '\ntrolley'
}
In the case of rank classification (letting the model select its the prediction the option with the highest log-likelihood), an example looks as follows:
{
'idx': [5, 0],
'inputs': [86, 8, 7142, 666, 6, 405, 8, 3, 834, 1518, 21, 19454, 42, 22227, 58, 19454, 744, 31, 17, 2112, 4553, 17742, 7, 12, 1953, 6, 298, 22227, 966, 373, 405, 5, 3, 834, 19, 72, 952, 12, 619, 16, 3, 9, 17742, 3298, 5],
'inputs_pretokenized': "In the sentence below, does the _ stand for Kyle or Logan?\nKyle doesn't wear leg warmers to bed, while Logan almost always does. _ is more likely to live in a warmer climate.",
'is_correct': True,
'targets': [19454, 1],
'targets_pretokenized': 'Kyle',
'weight': 1.0
}
To check all the prompted examples, you can use the Promptsource hosted tool and choose the Prompted dataset viewer mode in the left panel.
Data Fields
The data fields are the same among all splits:
answer_choices: the choices (in natural language) available to the modelinputs_pretokenized: the natural language input fed to the modeltargets_pretokenized: the natural language target that the model has to generateinputs: the tokenized input with T5's tokenizertargets: the tokenized target with T5's tokenizeridx: identifier of the (example, answer_option_id) in the case of rank classificationweight: a weight for the example produced by seqio (always set to 1.0 in practise)is_correct: whether the (example, answer_option_id) is the correct one
Data Splits
The list of data splits and their respective sizes is very long. You'll find the whole list in this file.
Dataset Creation
Curation Rationale
The Public Pool of Prompts relies on the Hugging Face Dataset library. Any public dataset in the Datasets library can be prompted. We select the datasets that have at least one subset in English and excluded datasets containing (predominantly) non-natural language examples.
We conservatively decided not to prompt datasets that contain potentially harmful content (for instance, datasets built on social media content). However, we sometimes prompt datasets that are purposefully built to measure bias and fairness of trained models, and reserve these prompted datasets (the validation or test sets) for evaluation purposes.
Source Data
Here's the full list of the datasets present in the materialized version of P3:
- Multiple-Choice QA
- CommonsenseQA
- DREAM
- QUAIL
- QuaRTz
- Social IQA
- WiQA
- Cosmos
- QASC
- Quarel
- SciQ
- Wiki Hop
- ARC
- OpenBookQA
- MultiRC
- PIQA
- RACE
- HellaSwag
- BoolQ
- Extractive QA
- Adversarial QA
- Quoref
- DuoRC
- ROPES
- SQuAD v2
- ReCoRD
- Close-book QA
- Hotpot QA
- Wiki QA
- Trivia QA
- Web Questions
- Structure-to-text
- Common Gen
- Wiki Bio
- Sentiment
- Amazon
- App Reviews
- IMDB
- Rotten Tomatoes
- Yelp
- Summarization
- CNN Daily Mail
- Gigaword
- MultiNews
- SamSum
- XSum
- Topic Classification
- AG News
- DBPedia
- TREC
- Paraphrase Identification
- MRPC
- PAWS
- QQP
- Natural Language Inference
- ANLI
- CB
- RTE
- Coreference Resolution
- WSC
- Winogrande
- Word Sense disambiguation
- WiC
- Sentence Completion
- COPA
- HellaSwag
- Story Cloze
Annotations
The prompts available in Promptsource are collected as part of BigScience, one-year long research workshop on large multilingual models and datasets. 36 contributors affiliated with 24 institutions in 8 countries participated to the prompt collection. Contributors are in majority machine learning researchers or machine learning engineers.
The main annotation guideline was that prompts needed to be grammatical and understandable by a native English speaker with no prior experience of the tasks. Additionally, prompts that required explicit counting or numerical indexing were removed in favor of natural language variants, e.g., instead of predicting indices of a span to extract (e.g. in extractive question answering), the model was expected to copy the span's text instead. With these minimal constraints, prompt writers were encouraged to use both formal and creative prompts and various orderings of the data. Most of the prompts correspond directly to a version of the original proposed task, although we also allowed prompts that permuted the original task (for instance, generating a document from its summary) or allowed for ambiguous output (for instance, not indicating a list of available choices).
The full annotation given to the contributors can be found here. *Note to self: the link is currently being updated with the)
Additional Information
Licensing Information
The dataset is released under Apache 2.0.
Citation Information
@misc{sanh2021multitask,
title={Multitask Prompted Training Enables Zero-Shot Task Generalization},
author={Victor Sanh and Albert Webson and Colin Raffel and Stephen H. Bach and Lintang Sutawika and Zaid Alyafeai and Antoine Chaffin and Arnaud Stiegler and Teven Le Scao and Arun Raja and Manan Dey and M Saiful Bari and Canwen Xu and Urmish Thakker and Shanya Sharma Sharma and Eliza Szczechla and Taewoon Kim and Gunjan Chhablani and Nihal Nayak and Debajyoti Datta and Jonathan Chang and Mike Tian-Jian Jiang and Han Wang and Matteo Manica and Sheng Shen and Zheng Xin Yong and Harshit Pandey and Rachel Bawden and Thomas Wang and Trishala Neeraj and Jos Rozen and Abheesht Sharma and Andrea Santilli and Thibault Fevry and Jason Alan Fries and Ryan Teehan and Stella Biderman and Leo Gao and Tali Bers and Thomas Wolf and Alexander M. Rush},
year={2021},
eprint={2110.08207},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
Contributions
Thanks to the contributors of promptsource for adding this dataset.